Re: Mathematica 5: Integration bug in quadratic found by the GEMM machine
- To: mathgroup at smc.vnet.net
- Subject: [mg46951] Re: Mathematica 5: Integration bug in quadratic found by the GEMM machine
- From: Bill Rowe <readnewsciv at earthlink.net>
- Date: Wed, 17 Mar 2004 02:29:05 -0500 (EST)
- Sender: owner-wri-mathgroup at wolfram.com
On 3/16/04 at 2:37 AM, akoz at mimuw.edu.pl (Andrzej Kozlowski) wrote: >Well, I hope that these are mostly "machine generated bugs". The >problem with this sort of approach of course is that you can take >just one genuine bug in, say, integration, and then by varying the >input in various ways generate 10, 100 or 10000 additional "bugs". >Whenever I hear of a "massive flow of bugs" I suspect something >like this at work. Could you clarify what you mean by "machine generated bugs"? If this was intended to mean a specific platform running Mathematica 5.0 then the particular error reported (Integrate[1/(-2 z^2 + 3 z - 1), {z, -1, 0}] resulting in Log[3/2]) is not specific to something running Windows since I get the same result with Mathematica 5.0 running on Mac OS 10.3.2 -- To reply via email subtract one hundred and four